Background

Drought stress symptoms can be qualitatively and quantitatively measured via image-based phenotyping. Imaging plant growth and the visual onset of stress symptoms results in greater sensitivity and data availability than what can be achieved from visual observations or endpoint analysis alone (Mutka and Bart, 2014; Tovar et al., 2018). Many protocols to measure plant stress via image analysis involve either a high up-front cost, a large experimental workspace, or integration of and proficiency with complex analytical software (Shakoor et al., 2017; Tovar et al., 2018). In contrast to these approaches, we have developed a simple, modular, low-cost protocol using a Raspberry Pi–controlled camera (which together require approximately $40 USD) to correctly evaluate drought resistance. We define drought resistance as any combination of plant strategies to cope with insufficient water availability required for maximum plant growth and reproduction, namely drought tolerance, drought avoidance, and drought escape (Blum, 2011; Deikman et al., 2012; Lawlor, 2013). Common plant responses to low water availability include increased root growth or maintenance of cell turgor via increased osmolyte production. Imaging plants as described in this protocol cannot specifically differentiate between drought avoidance and tolerance strategies. When paired with additional assays, however, both drought tolerance and avoidance can be quantified, as we recently demonstrated (Ginzburg et al., 2022). Drought escape (via earlier flowering time in response to reduced soil water content) can readily be quantified via time course imaging. This protocol requires no previous knowledge of software programming and requires only a small amount of space (approximately 1.5 m³ needed for 16 pots with Arabidopsis seedlings; Figure 4). While this protocol is suitable for any plant species, larger species or sample sizes may require additional space.

Some reports of image-based drought phenotyping platforms do not involve growing different genotypes together in the same pot or correcting for differences in % soil water content (SWC) between genotypes that are grown in separate pots (Kim et al., 2020). Experiments conducted without these crucial controls can yield potentially misleading results due to genotype-specific differences in water-use requirements (Lawlor, 2013). To account for these differences when evaluating drought resistance, soil water content for each genotype needs to be kept uniform throughout the drought treatment for all genotypes. This can be easily achieved, for example, by growing plants together in the same pot (Ginzburg et al., 2022), or, if in separate pots, by using soil moisture probes (Miralles-Crespo and van Iersel, 2011) or gravimetrically (Granier et al., 2006; de Ollas et al., 2019).

The persistence of reports that equate delayed onset of stress symptoms with drought resistance goes against established knowledge and slows progress in identifying true drought phenotypes (Lawlor, 2013). The protocol described here was recently used to differentiate between drought resistance and prolonged survival due to lower water usage in a dwarf Arabidopsis thaliana mutant chiquita1-1/cost1 (Ginzburg et al., 2022). Because plant responses to drought can vary depending on plant species, developmental stage, and degree of stress, there is no single set of measurements that can capture a drought phenotype in its entirety. Therefore, this protocol does not prescribe or outline in detail any specific measurements of plant stress. Rather, it provides readers with a foundation to properly set up a comparative drought experiment and capture images throughout the experiment to use for whichever downstream analyses are most appropriate to the experimental design and biological question(s). The straightforward methodology outlined here will be a valuable resource to the plant science community at large, not only for studying plant responses to drought but for any research requiring accurate visual comparisons of plant growth and development.